The Dartmouth Center for Clinical and Evaluative Studies has defined the geographical boundaries for hospital service areas (HSAs) for the United States using Medicare discharge data. HSAs have proven to be invaluable for three purposes: l) discovery and description of geographic variation in hospital resources and utilization, 2) testing hypotheses to explain the discovered variation in resources and utilization, 3) serving as a clustering variable for adjustment for correlated patients belonging to the same medical market area. Unfortunately the Medicare discharges upon which HSAs are based are almost exclusively derived from elderly patients. It has been assumed, but not demonstrated, that services areas for the elderly will suffice for pediatric data and research questions. However, compared to adults, children have a very different pattern of disease, different patterns of insurance coverage, and different patterns of hospitalization. The differences are probably even more dramatic in comparison to the elderly. Therefore research is needed to ascertain the feasibility and value of creating hospital service areas for pediatrics (HSAPs). For this exploratory project we propose to define and analyze HSAPs for a single large and diverse state, California. Our specific aims are to: l. Define HSAPs using the Dartmouth patient origin methods, 2. Determine the relative fit of Medicare-based HSAs and the new HSAPs to patient origin data, 3. Determine if HSAPs perform better than HSAs as a clustering variable for a typical child health services research question: What are the effects of race, bed supply and income on asthma discharge rate? If HSAPs are found to be effective they can lead to a number of new child health services research endeavors, including creation of an atlas of pediatric health care modeled on the Dartmouth Atlas of Health Care, studies of the stability of HSAPs over time, aggregation of HSAPs into hospital referral regions representing tertiary care, and improved analyses of racial and socioeconomic disparities in access to health care.